A South African commercial pilot with experience on twin engine aircraft has posted a comment on Avcom, a South African aviation forum. It's posted in the restricted area there, so I post the full text:
copy/paste:
"...I fly a twin-engined airliner. I have not flown an ATR or the Q400, both of which have full FADEC motors, but I have flown the DH8-300 in airline ops. All of these aircraft are certified to fly on one engine, in fact, they are certified to get airborne on one if the other (most critical) fails at V1. All airline training and operations assume that the aircraft will perform as advertised and that with good stick and throttle skills and sharp emergency procedures and CRM, you will climb away and live to fight another day. I have done this for 10 000 odd hours, and will never consider an off airport landing as long as we are still flying. After getting airborne in an airliner, the deck angle is such that you almost never see the ground in front of you, so the whole idea about planning for places to ditch is completely foreign to me. If, by some weird refueling error or bird strike, I happen to lose both engines, I have one of two plans in the CRJ. Anything above 2000' AGL in CPT or JNB, I can turn back for the other runway. Anything below that (or at places with single runways and up to 3000' (above 3000' I can turn back to same runway reciprocal)), I will look over the nose (which now points down for a change) and aim for anything open.
In other words, although there might be circumstances in which I will try an off airport landing, they are very specific and the spot I pick will be picked in the moment. These circumstances do not include a single engine failure.
Although the buildings loom large in the video, they are mostly less than 15 stories, so less than 200' high and I don't think the pilots even saw them with the nose so high. When I scoff at talk of avoiding buildings and planning to land on a highway or in a river, I don't do so because I want to denigrate the crew, but because that is what I know in my own heart and what other airline pilots think as well. We do things differently from recreational pilots because we have the opportunity to actually practice ("real") emergencies in the simulator. In April I will go for my 17th simulator recurrent in a CRJ200. With the initial, that adds up to over 200 hours in a level D simulator. Engine failures (even thrust reverser uncommanded deployment) is such old hat that we get marked on it without getting to practice it beforehand.
When I talk about ham-handed handling, it is also with understanding that all of us can let the aircraft get away from us and getting it back with little altitude available can be impossible. It does not change the fact that airline pilots are held to higher standards. We ask for more money because of it, but we have to produce the goods when required or our performance (or lack of it) will be called out by our peers. This attitude makes us all safer.
The normal climb schedule for the ATR72-600 (that is the newest model with full FADEC engines) calls for acceleration at a certain altitude (normally 400', but that can change with obstacle clearance limitations). You will get airborne at a speed somewhere between Vr and V2 and climb at V2 plus 10 or so kts until the accell altitude, where the power levers are pulled out of the T/o detent into the Climb detent. At this moment, you lose auto uptrim (more power from remaining engine if the other fails) and autofeather. It means that if an engine fails after this selection of the power levers (they apparently call them Power Management Selectors (PMS)), the crew are responsible to select more power (max power/max continuous power) and they have to manually feather the dead engine. Failing to do either or both will have significant consequences. The windmilling prop will have tremendous drag, because the CSU will immediately go full fine pitch to hold the RPM constant. Climb power would be somewhat less than full power on the live engine and with the nose high because of the climb, there will have to be an immediate and positive pitch forward to keep the speed from dropping to the wrong side of the drag curve or (God forbid) below Vmca or Vs.
Looking at the video, it seems as if flap is travelling from 0 to 15 in the last few seconds of the video. It also looks as if the left prop is in the process of being feathered. I cannot say what happened in that cockpit, but I do not think they had control of that aircraft from a few seconds after the engine failed. The fact that someone called "mayday", completely out of the normal sequence of events, I think supports my theory.
I have heard a few unsubstantiated rumours:
1. There were three pilots on board (Captain (maybe under training) 4000 odd hours; FO 6000 odd hours; someone on the jumpseat (maybe check/training captain) 14000 odd hours)
2. The particular aircraft had a history of engine problems with a very recent engine change.
3. The ATR has significant handling vices due to supercritical wing and stabilizer sections and small control surfaces.
(end of copy/paste)
Source:
http://www.avcom.co.za
and
http://www.avcom.co.za/phpBB3/viewtopic ... 0#p1555178